Kornia: how to apply morphological closing on 3D tensor?

1

Hi,

I want to apply morphological closing on a 3D grayscale/binary batched tensor. So far, i tried:

# importing modules
import torch
import torch.nn as nn
import torch.nn.functional as F
from kornia import morphology as morph

# creating random input 
input_tensor = torch.rand(1,1,8,8,8)
print(input_tensor[:,:,:,:,0])

# applying morphological closing
morph_kernel = torch.ones(3,3,3)
result = morph.closing(input_tensor, morph_kernel)
print(result[:,:,:,:,0])

But it leads to this error:

ValueError: Input size must have 4 dimensions. Got 5

Do you know a way to perform closing on 3D tensors (B,C,X,Y,Z) with kornia ? (or any other way working on tensors and allowing backpropagation)

Thanks